Senior Backend Engineer — TB-Scale AI Data Infra

constellation; San Francisco (CA); On-site; USD 180,000 - 230,000; Full time

Benefits offered by this job

Health insurance

Relocation assistance

SF office in Mission District

Team dinners & events

Parental leave (12 weeks)

Conference sponsorship

Job summary

Constellation is building an AI-human translation layer. We seek a Senior Software Engineer to own backend data collection, ingesting multimodal data at TB scale with a hybrid on-prem and cloud infra. Collaborate with AI teams to design high-performance data infrastructure for training-ready datasets.

You will architect fault-tolerant ingestion, multi-process apps, and device abstraction layers, while optimizing storage and network performance to support scalable data workflows.

Qualifications

• 3+ years of experience shipping production-grade backend systems.

Responsibilities

• Architect a fault-tolerant data ingestion engine
• Write robust multi-process applications for concurrent data streams
• Architect a unified device abstraction layer for heterogeneous peripherals
• Optimize I/O bound operations, and monitor bandwidth/latency
• Define storage topology for TB-scale daily ingestion and fast indexing/retrieval
• Manage provisioning and configuration of on-premise servers
• Build synchronization of terabytes of data from edge to cloud

Skills

Python/Rust/C++/Go

High-performance messaging (ZeroMQ/Kaf

Cloud infrastructure (AWS/GCP/Azure)

Data ingestion architecture

Video/Audio processing (FFmpeg/OpenCV)

Time-series DBs (TimescaleDB/InfluxDB/

Networking fundamentals

Parallel/multi-process development

Tools

ZeroMQ / Kafka / RabbitMQ / Redis Pub/Sub

AWS / GCP / Azure

FFmpeg / GStreamer / OpenCV

TimescaleDB / InfluxDB / ClickHouse

Job description

Constellation is building an AI-human translation layer. We seek a Senior Software Engineer to own backend data collection, ingesting multimodal data at TB scale with a hybrid on-prem and cloud infra. Collaborate with AI teams to design high-performance data infrastructure for training-ready datasets.

You will architect fault-tolerant ingestion, multi-process apps, and device abstraction layers, while optimizing storage and network performance to support scalable data workflows.